The Codex
Model character sheets
Exact Builds only. Stats include confidence grades and methodology version. Unscored launch Builds appear below until evidence is scored.
Code SmithArchitectPeermeth_v0_1_1
Claude Opus 4.8 · API · High effort (default)
The expensive architect — pays for hard repos, long agents, and fewer retries.
Overall 66.9STR 77.52026-07-11
Code SmithCollaboratorPeermeth_v0_1_1
Claude Sonnet 5 · API · High effort (default)
A polished code smith with expensive habits — still the sensible daily driver.
Overall 50.8STR 40.02026-07-11
Code SmithGeneralistPeermeth_v0_1_1
DeepSeek V4 Pro · API · Max effort
The open ceiling — near-frontier coding without the closed-model tax.
Overall 49.8STR 44.72026-07-11
MultimancerSagePeermeth_v0_1_1
Gemini 3.1 Pro · API · High thinking
The reasoning multimancer — 1M context, strong science, calibrated knowledge.
Overall 75.1STR 46.82026-07-11
GeneralistOperatorPeermeth_v0_1_1
GPT-5.6 Sol · API · Max effort
The ecosystem operator — max-effort Sol for hard agents, not cheap chat.
Overall 79.7STR 94.92026-07-11
Code SmithOperatorPeermeth_v0_1_1
Grok 4.5 · API · High effort
The alt-frontier code smith — near-Opus agentic coding at a third the token bill.
Overall 74.0STR 68.82026-07-11
Launch slate · not yet scored
Claude Sonnet · Consumer chat
bld_anthropic_claude_sonnet_consumer
DeepSeek V4 Flash · API
bld_deepseek_v4_flash_api
GLM · API
bld_glm_api
Gemini Flash · API
bld_google_gemini_flash_api
GPT Mini/Fast · API
bld_openai_gpt_mini_api
Qwen Coder · Local Q4
bld_qwen_coder_local_q4